130 28 34MB
English Pages 756 [737] Year 2018
ERGONOMICS HOW TO DESIGN FOR EASE AND EFFICIENCY THIRD EDITION Katrin E. Kroemer Elbert Henrike B. Kroemer Anne D. Kroemer Hoffman
Academic Press is an imprint of Elsevier 125 London Wall, London EC2Y 5AS, United Kingdom 525 B Street, Suite 1650, San Diego, CA 92101, United States 50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom Copyright © 2018 Elsevier Inc. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the Publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions. This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein). Notices Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library ISBN: 978-0-12-813296-8 For information on all Academic Press publications visit our website at https://www.elsevier.com/books-and-journals
Publisher: Matthew Deans Acquisition Editor: Brian Guerin Editorial Project Manager: Thomas Van Der Ploeg Production Project Manager: Vijay Bharath. R Designer: Mark Rogers Typeset by Thomson Digital
Preface to the Third Edition
What do we know about the human body and mind at work? Given what we know, how then should we design the work task, tools, the interface with the machine, and work procedures so that the human can perform safely, efficiently, and with satisfaction--perhaps even enjoy working? These challenges are the main themes of this book. The solutions are the WHY and HOW of ergonomics/human engineering.
These first lines in the 1994 first edition of Ergonomics–How To Design For Ease And Efficiency appeared again in the second edition of 2001, amended 2003. They remain the motto of this third edition of our book. We have updated and refreshed content to reflect and/or clarify current approaches to mind and body, including more recent findings, research, applications, and equipment. In particular, we’ve substantially updated the chapters on the computer workplace (Chapter 10) and designing for special populations (Chapter 14) and added a chapter on the individual within an organization (Chapter 11). Further, we have made improvements to the organization and flow of the information in the chapters. For ease of reading, we avoided interrupting the text with names and dates of other authors but have used a small marker to refer to further “Notes” at the end of each chapter for the interested reader. We have left out some older references which, in the earlier editions, supported ergonomic findings and recommendations that are now well established and generally accepted. Of course, we continue to provide the complete references for all previously and newly included material, so the third part of the book remains a substantive source of past and present ergonomic information. We remain grateful to all who gave suggestions for improvements. Our principal influence is our father, Professor Emeritus Karl Heinrich Eberhard Kroemer, who is the original primary author of this text and remains our inspiration for all things ergonomic. His passion for relevant and practical research, rigorous engineering, and dissemination of information is reflected in this book. He has provided significant updates, insight, and guidance, and is our role model for methodical technical work with an eye for improving the human work condition. We would like to hear from you! Please, tell us what we did well, what we should do better in the next edition, and remind us of any past knowledge not to be forgotten. We can be reached at [email protected]. Katrin E. Kroemer Elbert, PhD Henrike B. Kroemer, PhD, LP Anne D. Kroemer Hoffman, MBA
ix
Using This Book
In this book, we discuss interactions of the human with work tasks, task equipment and the work environment. Our intent in exploring the interactions is to build a knowledge-based understanding so that we can: • • • • •
utilize human abilities, amplify human capabilities, facilitate human efficiency, avoid overloading or underloading, and, we hope, enjoy doing our tasks.
This book comprises physical, psychological, biomechanical, and social features of the human as they concern individual or group relations to job, equipment, task environment, and organization. The authors include administrative concerns, consider economic constraints, and reflect aspects of today’s ever-evolving technology. This book is comprehensive of all facets of human factors/ ergonomics. The writing and layout of this book rely on the concept of providing basic background knowledge and building from there to detailed information that allows the reader to delve further into special and novel applications. The person new to the field, the undergraduate and the graduate student will all gain a broad and thorough understanding of “Human-Technology Systems.” The authors acknowledge that, although the physical aspects of humans are changing only slowly, requirements for design are ever changing as the ergonomic database is ever expanding. To that end, we include general references to international standards and organizations to allow the interested reader to seek the latest information and requirements, which are updating more frequently than this text. Further, we intend this book to be useful to any interested reader and try to present issues relevant to all humans. We use the pronouns “he” and “she” interchangeably when referring to gender nonspecific features of human factors. While we understand that “gender” may be used to describe masculinity/femininity, much of the literature we review assumes gender refers to biological aspects of maleness/femaleness. Ergonomic understanding will benefit established professionals, both specialists and generalists, concerned with people’s performance and well-being at work. Among these are the designer, engineer, architect, industrial hygienist, industrial physician, industrial psychologist, occupational nurse, and manager. Of course, all of us are interested in humanizing work, making it more safe, efficient, and satisfying.
xi
xii
Using This Book
EASY READING To avoid breaking the flow of reading, we have placed small markers in the text: they indicate where references or explanations are in order. These appear in a separate “Notes” section at the end of each chapter, which the reader may skip or consult.
THREE SECTIONS OF THIS BOOK Section I: “The Ergonomic Knowledge Base” consists of Chapters 1–7. Here, we explore the properties of the human body and mind in the task environment. The focus is on human dimensions, capabilities and limitations; that is, the human factors to be considered in designing for ease and efficiency. For everybody’s convenience, we use traditional disciplinary divisions into anatomy, physiology and psychology, whereas, as we all know, the human functions holistically and synergistically. The chapters contain references to overlapping information covered in other chapters. Section II: “Design Applications” contains Chapters 8–15. Here, we discuss the design of tasks, equipment and environment based on the knowledge base developed in Section I about human attributes, strengths and limitations. Section III: “Further Information” at the end of the book, includes a list of all References mentioned in the text (including previous editions), an extensive Glossary with concise descriptions and definitions, and a detailed Index that refers the reader to specific pages.
THREE WAYS TO USE THIS BOOK 1. Read straight through from beginning to the end, as a university course, and work on the “Challenges” listed at the end of each chapter. 2. Select a chapter of particular interest, absorb the background information, and proceed to design applications that make use of that information. 3. Start with the index, pick a topic of interest, and look up the information in the book section(s) that the index cites.
Introducing Ergonomics and Human Factors Engineering WHAT ERGONOMICS IS Ergonomics is the use of scientific principles, methods, and data drawn from a variety of disciplines for the development of systems in which people play a significant role. The field of applications extends from a single person using a simple tool to a complex multi-person socio-technical organization. Ergonomic specialists rely on the understanding that systems are meant to benefit people, whether they are consumers, production workers, system operators or maintenance crews. This usercentered design philosophy acknowledges human variability as a design parameter. The resultant design features utilize human capabilities, consider human limitations and have built-in safeguards to avoid or reduce the effects of human error or of system failure. Ergonomics is neutral: it takes no sides, neither employers’ nor workers’. It is not for or against progress. It is not a philosophy, but a scientific discipline and practical technology.
WHAT ERGONOMICS DOES Ergonomics focuses on the human as the most important component of our technological systems. This explains the early terms “human factors engineering,” “man-machine systems,” and the recent “human-systems integration.” The aim is to assure that all human-made tools, devices, equipment, machines, environments and their organizations advance, directly or indirectly, the well-being of humans and their performance. Accordingly, ergonomics has three related tasks: 1. Study, research, and experimentation to determine specific human traits and characteristics that need to be known for engineering design. 2. Apply this knowledge in the design of tools, machines, shelter, environments, work tasks and procedures so that they fit and accommodate the human. 3. Observe the actual performance of humans and their equipment in the real or a simulated environment, assess the suitability of the designed human-machine system and determine whether improvements are possible.
HOW ERGONOMICS CAME ABOUT Using objects found in the environment as tools is an ancient activity. Pieces of stone, bone, and wood must have been selected for their fit to the hand and their suitability as cutters, scrapers, pounders, and missiles. Purposeful shaping of these tools naturally came next, followed by assembling raw
xiii
xiv
Introducing Ergonomics and Human Factors Engineering
materials to make, for example, protective clothing and shelters. These fundamental “ergonomic” activities grew into individual and communal skills and then into processes for creating finished products from diverse materials. As human society grew more complex, organizational and managerial challenges developed. Training workers and soldiers, for example, became necessary, together with forming and controlling their behavior. Major projects, such as building the pyramids of ancient Egypt, assembling armies for warfare, sheltering the inhabitants of cities and supplying them with food and water, required sophisticated knowledge of human needs and desires; careful planning and complex logistics had to be mastered. The aims and means of training became sophisticated as well: Roman soldiers, for example, underwent well-organized exercises and conditioning until they could perform military exercises with ease: “drying the legions” of the Roman Empire relied on the principle of adapting and improving the physiological capabilities of the recruits to meet challenging physical requirements; when they no longer showed sweat on the skin, they were “dry” and fit.
Evolution of Disciplines Artists, military officers, employers and athletes apparently were always interested in body build and physical performance. Specialized “medicine men” and “herb women” treated illnesses and injuries. About 400 BC, Hippocrates (often considered the father of western medicine) described a scheme of four body types: the “moist” type was believed to be dominated by black gall, the “dry” type by yellow gall, the “cold” type by slime, and the “warm” type by blood. These “humors” had been in some balance to keep an individual healthy. Knowledge of medicine, anatomy and anthropology began to assemble. Over the centuries, more exact information accumulated into specialized disciplines. In the 15th– 17th centuries, gifted people such as Leonardo da Vinci (1452–1519) and Alfonzo Giovanni Borelli (1608–1679) could still master all existing knowledge of anatomy, physiology and equipment design; these individuals were artist, scientist, and engineer in one. Da Vinci, along with being an exceptional artist, understood nature, mechanics, anatomy, physics, architecture, and weaponry. His designs for machines included a bicycle, helicopter, and an airplane. Borelli’s book de Motu Animalium, published in 1680, applied the modern principle of scientific investigation by continuing Galileo’s custom of testing hypotheses against observation. Borelli extended rigorous analytical methods to biology (he can be called the “father of biomechanics”); he was the first to understand that the levers of the human musculoskeletal system favor quick motion rather than force. In the 18th century, the sciences of anatomy and physiology diversified and amassed specific detailed knowledge. Psychology began to develop as a distinct field of study. Well into the 19th century, the sciences tended to be oriented toward theories: the stereotype is the scientist in a white coat who devotes his life to research in the laboratory. But increasing industrialization with its employment of human workers, focused interest on “applied” aspects of the “pure” sciences.1 In the early 1800s, in France, Lavoisier, Duchenne, Amar, and Dunod researched energy capabilities of the human body. Marey developed methods to describe human motions at work. Bedaux made studies to determine work and payment systems before Taylor and the Gilbreths did similar work in the United States in the early 1900s. In England, the Industrial Fatigue Research Board considered theoretical and practical aspects of the human at work. In Italy, Mosso constructed dynamometers and ergometers to research fatigue. In Scandinavia, Johannsson and Tigerstedt developed the scientific disciplines of work physiology. In 1913, Rubner founded a Work Physiology Institute in Germany.
Introducing Ergonomics and Human Factors Engineering
xv
In the United States, Benedict and Cathcard described the efficiencies of muscular work in 1913. The Harvard Fatigue Laboratory was established in 1927. In the first half of the 20th century, applied physiology and psychology were well advanced and widely recognized, both in their theoretical research “to study human characteristics” and in the application of this knowledge “for the appropriate design of the living and work environment.” Two distinct approaches to studying human characteristics had developed: one concerned chiefly with physiological and physical properties of the human, the other interested mainly in psychological and social traits. Although there was much overlap between these approaches, the physical and physiological aspects were studied mainly in Europe and the psychological and social aspects and North America. Directions in Europe Based on a broad fundament of anatomical, anthropological and physiological research, “work” physiology assumed great importance in Europe, particularly in the hunger years associated with the First World War. Marginal living conditions stimulated research on such topics as the minimal nutrition required to perform certain activities; the consumption of energy while doing agricultural, industrial, military and household tasks; the relationships between energy consumption and heart rate; the assessment of muscular capabilities; suitable body postures at work; the design of equipment and work stations to fit the human body. Another development in the 1920s was psycho-technology, which involved testing individuals for their ability to perform physical and mental work, their vigilance and attention, their ability to carry mental workload, their behavior as drivers of vehicles, their ability to read road signs, and related topics. Directions in North America Most psychologists around 1900 were strictly scientific and deliberately avoided studying problems that strayed outside the boundaries of pure research. Some investigators, however, pursued practical concerns, such as sending and receiving Morse code, measuring perception and attention at work, using psychology in advertising, and promoting industrial efficiency. A particularly important step was the development of intelligence testing,2 used to screen military recruits during the First World War and, later, to select industrial workers for their mental capabilities which certain jobs required. The concept of industrial psychology won acceptance. Some of the best-known, most puzzling findings in industrial psychology resulted from experiments at the Hawthorne Works near Chicago in the mid-1920s. The experiments were designed to assess relationships between efficiency and lighting in workrooms where electrical equipment was produced. The bizarre finding was that the workers’ productivity increased whenever the illumination was changed, regardless of increase or decrease; apparently, productivity responded to the attention paid to the workers by the researchers. This phenomenon became known as the observer or Hawthorne Effect.3 Industrial psychology4 divided into special branches, including personnel psychology, organizational behavior, industrial relations, and engineering psychology. Under the pressures of the Second World War, the “human factor” as part of a “man-machine system” became a major concern. Technological development lead to machines and systems that put higher demands on the attention, endurance, and strength of individuals and teams than many could muster. For example: operators had to observe radar screens over periods of many hours, with the intent of detecting and distinguishing some blips from others. In high-performance aircraft, the pilot had to endure forceful accelerations,
xvi
Introducing Ergonomics and Human Factors Engineering
for instance in tight turns and in steep dives or climbs. In these cases, the pilot might be unable to operate hand controls properly and could even black out. Crew members had to fit into tiny tank and aircraft cockpits, and (from the 1960s on) into spacecraft, which required that small persons be selected from expert aviators. Stressful conditions make it difficult to maintain morale and performance. New tasks and machinery generate new needs to consider human physique and psychology, purposefully, and knowingly, in the design of jobs, equipment, and environments.
Names for the Discipline: “Ergonomics” and “Human Factors” Early on, in Europe and North America, anthropologists, physiologists, psychologists, sociologists, statisticians, and engineers used various terms to describe their activities of studying the human and applying the information obtained in design, selection, and training. In January of 1950, British researchers met in Cambridge, England, to select a name for the new society to represent their activities. Among others, the term “ergonomic” was proposed.5 In late 1949, KFH Murrell had derived that word from the Greek terms ergon, indicating work and effort, and nomos, meaning law or usage; apparently, he re-invented a word already used by W. Jastrzebowski in Poland nearly a hundred years earlier. That term was neutral; it implied no priority among contributing disciplines; it was easily remembered and recognized and could be used in any language. “Ergonomics” was formally accepted as the name of the new society at its council meeting in early 1950.6 In the United States, a group convened in 1956 to establish a formal society. They selected “human factors” instead of “ergonomics.” Often, the word “engineering” is added or substituted to indicate applications, as in “human (factors) engineering”. In 1992, the Human Factors Society renamed itself the Human Factors and Ergonomics Society7 HFES with (in 2016) about 4500 individual members.8 The International Ergonomics Association IEA has 54 member societies (in 20169) with 28 in Europe, 13 in Asia, 9 in the Americas and 4 in Africa and the Oceania regions. There has been some discussion of whether human factors differs from ergonomics—whether one relies more heavily on psychology or on physiology, or is more theoretical or practical than the other. Today, the two terms are usually considered synonymous: the Canadian Society uses “human factors” in its English name, and “ergonomie” in its French version.
TODAY’S ERGONOMIC KNOWLEDGE BASE The development of space travel first forced humans to fold themselves into minuscule capsules; now people live and function in near weightlessness for months. Today, many professional activities involve long hours using computers of one kind or another as the primary work tools. Yet, heavy physical work still persists in many jobs in industry, in commercial fishing, in agriculture and forestry—some hard work is even new, such as in airline baggage handling. The field of ergonomics/human factors continues to grow and to change, driven by new technologies and by the resulting new tasks for people. Classic sciences still provide fundamental information about human beings (Fig. 1). The anthropological basis consists of Anatomy, describing the build of the human body; Orthopedics, concerned with the skeletal system; Physiology,
Introducing Ergonomics and Human Factors Engineering
xvii
FIGURE 1 Origins, developments, and applications of ergonomics/human factors.
dealing with the functions and activities of living body, including the physical and chemical processes involved; Medicine, concerned with injuries and illnesses and their prevention and healing; Psychology, the science of mind and behavior; and Sociology, concerned with the development, structure, interaction, and behavior of individuals or groups. Of course, physics, chemistry, statistics, and mathematical modeling supply methodology and knowledge. Several applied disciplines developed from these basic sciences; these include primarily Anthropometry, the measuring and description of the physical dimensions of the human body; Biomechanics (Anthromechanics), describing the physical behavior of the body in mechanical terms; Industrial Hygiene, concerned with the control of occupational health hazards; Industrial Psychology, discussing people’s attitudes and behavior at work; Management, dealing with and coordinating the intentions of the employer and the employees; and Work Physiology, applying physical knowledge and measuring techniques to the body at work. Of course, associated disciplines such as Labor Relations have developed which are also part of, or contribute to, or overlap with, ergonomics. Their topical areas overlap and intertwine; their research produces practical applications in such fields as Industrial Engineering, also called Integrated Engineering, concerned with the interactions among people, machinery, and energies; Bioengineering, working to replace worn or damaged body parts; Systems Engineering, in which the human is an important component of the overall work unit; Safety Engineering and Industrial Hygiene, which focus on the well-being of humans; and Military Engineering, which relies on the human as soldier or operator. As computers have become integrated into all aspects of modern day life, computer-aided design is used in all aspects of engineering. Naturally, other application disciplines also rely on ergonomic knowledge and data, such as oceanographic, aeronautical, and astronautical engineering.
xviii
Introducing Ergonomics and Human Factors Engineering
GOALS OF ERGONOMICS The (US) National Research Council asserted in 198310 that “design begins with an understanding of the user’s role in overall system performance and that systems exist to serve their users, whether they are consumers, system operators, production workers, or maintenance crews. This useroriented design philosophy acknowledges human variability as a design parameter. The resultant designs incorporate features that take advantage of unique human capabilities as well as built-in safeguards.” The human is never indentured to the system but is the beneficiary, the passenger, the participant, the operator, the supervisor, the controller, and the decision maker. There is a hierarchy of goals in ergonomics. The most essential and basic task is to generate tolerable working conditions that do not pose unavoidable dangers to human life or health. When this basic requirement is assured, the next goal is to generate acceptable conditions to which the people involved can voluntarily agree. The final goal is to generate optimal conditions that are so well adapted to human characteristics, capabilities, and desires that physical, mental, and social well-being is achieved.
ERGONOMICS DEFINED Ergonomics (human factors, human engineering, human factors engineering) is the study of human characteristics for the appropriate design of the living and work environment. Its fundamental aim is that all man-made tools, devices, equipment, machinery, and environments advance, directly or indirectly, well-being and performance of people. As more knowledge about humans becomes available, as novel opportunities develop to apply human capabilities in modern systems, and as new needs arise for protecting the person from outside events, ergonomics changes and expands.
NOTES 1 For more details, see McFarland (1946), Chapanis, Garner, and Morgan (1949), Lehmann (1953, 1962), Floyd and Welford (1954), Woodson (1954), Brouha (1960, 1967), Grandjean (1963), and Scherrer (1967). 2 Gould (1981) provides a partly amusing, partly disturbing account of the early years of such testing. 3 Roethlisberger and Dickson (1943), Parsons (1974, 1990), and Jones (1990). 4 The term first appeared as a misprint of “individual” psychology, as Muchinsky (1983), p. 12. 5 Edholm and Murrell (1974), Monod and Valentin (1979), and Koradecka (2000). 6 The original proposal for the name included two alternative suggestions. One was “Ergonomic Society.” Note that there is no “s” following Ergonomic; the final “s” apparently slipped into the ballot and has made deriving an adjective or adverb difficult. The other alternative was “Human Research Society.” 7 Christensen, Topmiller and Gill (1988), Kroemer (1993a,b), and Meister (1999). 8 Accessed 7 July 2016 from https://hfes.org. 9 Accessed 21 June 2016 from https://iea.cc/. 10 National Research Council, Committee on Human Factors (1983), Research needs for human factors, pp. 2–3, Washington, DC: National Academy Press.
C H A P T E R
1
Size and Mobility of the Human Body O U T L I N E 1.1 Humans Spreading Over the Earth............................................................................................................. 4 1.2 Anthropology and Anthropometry.............................................................................................................5 1.2.1 Measurement Techniques..................................................................................................................6 1.2.2 Body Posture during Measurements...............................................................................................7 1.2.3 Body Measures..................................................................................................................................... 8 1.2.4 Classical Measuring Techniques.......................................................................................................9 1.2.5 New Measurement Techniques......................................................................................................10 1.3 Available Anthropometric Information.................................................................................................... 11 1.3.1 Variability............................................................................................................................................11 1.3.2 Population Samples...........................................................................................................................15 1.4 Anthropometric Statistics............................................................................................................................ 18 1.5 Using Anthropometric Data........................................................................................................................ 23 1.6 Body Proportions........................................................................................................................................... 25 1.6.1 Body Image.........................................................................................................................................25 1.6.2 “Desirable” Body Weight..................................................................................................................26 1.6.3 Body Mass Index................................................................................................................................ 26 1.7 Dealing With Statistics.................................................................................................................................27 1.8 How to Obtain Missing Data...................................................................................................................... 31 1.9 “Fitting” Design Procedures........................................................................................................................ 34 1.9.1 Safe Distances.................................................................................................................................... 39 1.10 Design Steps................................................................................................................................................... 39 1.11 Chapter Summary......................................................................................................................................... 42 1.12 Challenges...................................................................................................................................................... 42
Ergonomics http://dx.doi.org/10.1016/B978-0-12-813296-8.00001-3
3
Copyright © 2018 Elsevier Inc. All rights reserved.
4
1. Size and Mobility of the Human Body
Overview “How tall are you? What’s your weight?” For us to design things that fit people, we must know the dimensions of their bodies; however, body builds can vary considerably among individuals. Measurements of body sizes are available for some populations, but for many people on Earth only estimates exist. If we have suitable statistical descriptors and know the relationships among various body data, we can calculate probable body dimensions; if not, we must measure. Understanding the properties, capabilities, and limitations of the body allows us to design equipment and tools that utilize and enhance human strengths.
1.1 HUMANS SPREADING OVER THE EARTH Development of humans
Homo in Africa
Neanderthals and Cro-Magnons
Humans spread from Africa
Emigrants from Asia in the Americas
The human species appears to have grown like a big bush: some branches at first develop but then dry up, while others grow more and more twigs, some of which may vanish while others flourish. We can trace the development of the human race by fossils and by reconstructing mitochondrial DNA over several million years in Africa, for hundreds of thousands of years in Europe and Asia, and for some 20,000 years in the Americas. Current paleoanthropology suggests that Australopithecine was a predecessor (or possibly close evolutionary cousin) of the genus Homo about 3 million years ago in Africa, where Homo erectus then developed. One humanoid branch started about 250,000 years ago and remained in Africa. Another branch developed 60,000 or 70,000 years later. Some of its members stayed in Africa, others spread into Eurasia. Eventually, Homo sapiens spread all over the Earth.1 Remains of anatomically modern humans who lived 130,000–180,000 years ago have been found in South Africa and in the Levant. About 150,000 years ago, the Neanderthals emerged, primarily in central Europe. They apparently were stocky, heavy-set, and cold-adapted with a brain as big as our current one. For several thousand years, they existed side-by-side with Cro-Magnons but then vanished about 30,000 years ago. The Cro-Magnons grew into Homo sapiens. Popular notions about the different appearances of Cro-Magnons and Neanderthals are mostly based on conjecture, often in the style of a commercial Hollywood-type movie. For example, there is no indication that the Cro-Magnons were dark skinned, or the Neanderthals light. Furthermore, there is no evidence of violent struggles for superiority between the two races.2 From Africa, Homo sapiens spread over the Earth. Roughly 50,000 years ago, Australia was settled by early humans who arrived from eastern Indonesia. Their descendants became the Aboriginal population. Most of the current inhabitants of Indonesia, the Philippines, and parts of Southeast Asia may be descendants of a population that emigrated from modern-day Taiwan about 4000–6000 years ago.3 Waves of peoples, the earliest around 20,000 years ago, crossed what was then the Bering land bridge from East Asia to modern-day Alaska. Some moved into the areas of today’s Canada and the United States, others followed the pacific coastal areas into
I. The Ergonomic Knowledge Base
5
1.2 ANTHROPOLOGY AND ANTHROPOMETRY
South America. Their descendants populated the entire hemisphere, becoming the ancestors of North, Central, and South American native peoples. After its old history of Neanderthals and Cro-Magnons, the European population was reconstituted twice fairly recently: around 8000 years ago by people from the Near East and then 2000 years later by Indo-Europeans from southern Russia. Thus, the human stock with its many current branches appears African in origin and about a quarter-million years old. Today, the number of people is growing fast; “population explosions” are occurring in some parts of the Earth. The total number of humans was about 109 (1000 million, or 1 billion) around 1800. In 1900, about 1.7 billion people lived on Earth. The second billion was reached by 1930. The third billion was present in 1960, the fifth in 1987. In 1998, about 5.8 billion people lived on Earth, including 4.6 billion in developing countries. In 2017, the total human population on Earth has been estimated to be 7.5 billion. If current birth and death rates continue, about 10 billion people will live on Earth in 2050; projections beyond this time vary widely. Emigration from certain areas and immigration to others are on a much smaller scale than population growth but can be locally of great importance. In North America, for example, during the last few centuries, waves of immigrants from certain geographical areas have been changing the composition of the inhabitant population, replacing most native peoples with Europeans. In today’s United States, the influx of Cubans and Haitians is strongly felt in Florida, the arrival of Central and South Americans affects mostly southwestern states, and Asians are very evident along the Pacific coast.
Europe
Population growth
Local population changes
1.2 ANTHROPOLOGY AND ANTHROPOMETRY Anthropology, the study of mankind, was primarily philosophical and aesthetic in nature until about the middle of the 19th century. Size, proportions, and appearance of the human body have always been of interest to artists, warriors, and physicians. In the mid1800s, Adolphe Quetelet applied statistics to anthropological data of body measurements. This was the beginning of modern anthropometry, the measurement and description of the human body. By the end of the 19th century, anthropometry was a widely applied scientific discipline, used both in measuring the bones of early people and in assessing the body sizes and proportions of contemporaries. A new offspring, biomechanics (related to structure and movement), emerged. Today, engineers have become highly interested in the application of anthropometric and anthromechanical information.4 Unification of measuring methods became necessary and was achieved primarily by anthropologists who convened 1906 in Monaco and 1912 in Geneva. They established landmarks on the body, mostly on bones, to and from which to take measurements. In 1914, Rudolf Martin published his Lehrbuch der Anthropologie. The several editions of this authoritative textbook shaped the discipline for decades. Beginning in the 1960s, increasing engineering needs for anthropometric information, newly developing measuring techniques, and advanced statistical considerations stimulated the need for updated standardization. Since the 1980s, the International Standardization Organization (ISO) has established conforming anthropometric techniques worldwide.5
I. The Ergonomic Knowledge Base
Measurement of the human body
Standardized measurements
6
1. Size and Mobility of the Human Body
FIGURE 1.1 Reference planes in anthropometry. Note the directions of x, y, and z—see text.
1.2.1 Measurement Techniques
Reference planes
Body length
Most body measurements are defined by the two endpoints of the distance measured. For example, forearm length is the elbow-to-fingertip distance; stature (height) starts at the floor on which the subject stands and extends to the highest point on the skull. Fig. 1.1 shows the three standard reference planes: the medial (mid-sagittal), the frontal (or coronal), and the transverse planes, all at 90 degrees to each other and usually set to meet in the center of mass of the whole body. A Cartesian coordinate system is seldom employed in anthropometry but used routinely in anthromechanics (see Chapters 2 and 6)6; the convention is to set the +x axis pointing forward from the subject, +y to the subject’s left, and +z upward. The origin of the axes may be moved to any point of the body, such as to a hand, if appropriate for modeling. The classic reference posture of a person standing upright is similar to that shown in Fig. 1.1 but with the heels of the feet kept together, called “anatomical position.” Fig. 1.2 lists other descriptive terms.
I. The Ergonomic Knowledge Base
7
1.2 ANTHROPOLOGY AND ANTHROPOMETRY
FIGURE 1.2 Terminology used in anthropometry. Source: Adapted from Kroemer, Kroemer, and Kroemer (2010).
For measurement of stature, the subjects may assume one of four customary positions: standing naturally upright; standing stretched to maximum height; leaning against a wall with the back flattened and buttocks, shoulders, and back of the head touching the wall; or lying on the back. Lying supine results in the largest measure; it is mostly used with infants who are unable to stand. The difference between measures when the standing subject either stretches or just stands upright can easily be 2 cm or more. This demonstrates that standardization is needed to assure uniform postures and comparable results.7
1.2.2 Body Posture during Measurements For most measurements, the subject stands upright with body segments in line with each other or at right angles. When measurements are taken on a seated person, the flat I. The Ergonomic Knowledge Base
Body posture
8
1. Size and Mobility of the Human Body
FIGURE 1.3 The Ear–Eye Line describes head posture in the lateral view. The E–E Line is also the reference for the angle of the line of sight, LOSEE, in the medial plane.
Head posture
and horizontal surfaces of feet and foot support are so arranged that the thighs are horizontal, the lower legs vertical, and the feet flat on their horizontal support. The subject is nude, or nearly so, and does not wear shoes. If measurements include the head, it is positioned upright (“erect”), with the pupils on the same horizontal level. The traditional term for this posture was “head in the Frankfurt Plane,” but an easier description is by the Ear–Eye (E–E) Line: it passes through the right ear hole and the outside juncture of the lids of the right eye. When the head is upright, the E–E Line is angled by about 11 degrees above the horizon, as shown in Fig. 1.3.
1.2.3 Body Measures Height is a straight-line, point-to-point vertical measurement. Breadth is a straight-line, point-to-point horizontal measurement running across the body or a segment. Depth is a straight-line, point-to-point horizontal measurement running fore-aft the body. Distance is a straight-line, point-to-point measurement between landmarks on the body. Curvature is a point-to-point measurement following a contour; this measurement is usually neither closed nor circular. Circumference is a closed measurement that follows a body contour; hence this measurement is usually not circular. Reach is a point-to-point measurement following the long axis of the arm or leg. All measurements are taken on people maintaining prescribed erect standing or sitting stances.
I. The Ergonomic Knowledge Base
9
1.2 ANTHROPOLOGY AND ANTHROPOMETRY
FIGURE 1.4 Anatomical landmarks in the sagittal view.
Figs. 1.4 and 1.5 illustrate anatomical landmarks on the human body which serve as starting and endpoints for body measures.
1.2.4 Classical Measuring Techniques The Morant technique used a set of measurement grids, usually attached to the inside corner of two vertical walls meeting at right angles. The subject was placed in front of the grids, and projections of the body onto the grids were used to determine anthropometric values. Related box-like jigs with grids are still in some use for determining foot dimensions. In the classical method, the anthropometrist’s hands guide special measuring instruments to the bony landmarks on the body of the subject. The largest instrument is the anthropometer, a graduated rod with a sliding edge at right angle, used to measure long I. The Ergonomic Knowledge Base
Grid technique
Hand-held measuring devices
10
1. Size and Mobility of the Human Body
FIGURE 1.5 Anatomical landmarks in the frontal view.
Simple but clumsy
straight distances. The spreading caliper consists of two curved branches joined in a hinge, used to determine the distance between the tips of the branches. A small sliding caliper serves to take short measurements, such as finger thickness or finger length. Thickness of skinfolds, a measure of body fat, is determined with a skinfold caliper. A cone serves to measure the diameter around which fingers can close. Circumferences and curvatures are measured with a tape. Measuring body dimensions with these traditional instruments is simple in principle, but requires experience and skill and consumes a great deal of time. Each measurement and tool must be selected in advance, and what was not measured in the test session remains as an unknown. A major shortcoming of the classical technique is that most body measures appear unrelated to each other in space. For example, as one looks at a subject from the side, stature, eye height, and shoulder height are located in different yet undefined frontal planes.
1.2.5 New Measurement Techniques Photography
Photographs can record all three-dimensional (3D) aspects of the human body. They allow the recording of practically infinite numbers of measurements, taken from the I. The Ergonomic Knowledge Base
11
1.3 AVAILABLE ANTHROPOMETRIC INFORMATION
recording at one’s convenience. However, photographs also have drawbacks: the body is depicted in two dimensions; a scale may be difficult to establish; parallax distortions occur; and bony landmarks under the skin cannot be palpated on the photograph. For these and other reasons, two-dimensional photographic and video anthropometry have not been widely used. An electromechanical probe8 can be placed on body landmarks and their locations can be registered in three dimensions. Computerized data storage and processing allows 3D identification of body points and of their changes by motion. A laser can be used as a distance-measuring device to determine the shape of irregular bodies. Measurements are taken by either rotating the body to be measured, or by rotating the sending and receiving units of the laser device around the body. Markers may be placed on points of the body surface so that the laser can recognize them, for example to indicate the location of a bone landmark. Scanning9 is fast and collects vast amounts of measurements. Computerized data storage and processing allows detailed 3D descriptions of the body and of changes due to motion, training, or aging.
Placing a probe
Body scanning
1.3 AVAILABLE ANTHROPOMETRIC INFORMATION In the past, interest in the body build of populations other than one’s own group was based mostly on curiosity and general wish to know; however, as industry and marketing began to reach around the globe, body size became a matter of practical interest to designers and engineers. Around 1970, first compilations of worldwide anthropometric information appeared.10 Since then, an increasing number of publications describe national populations. Modern measuring technology and sophisticated statistics allow anthropometric assessment of populations to be taken rapidly and accurately. Table 1.1 shows 1990 estimates of stature of the Earth’s population, divided into regional groups. Table 1.2 contains measured data on specific population samples. These data are included only to demonstrate the variability that may be encountered, and not to reflect current populations: in many cases the surveys were done decades ago and included only small groups.11
Anthropometric sourcebooks
Earth’s population
1.3.1 Variability Anthropometric data show considerable variability stemming from four sources: variability in the measurement itself, variability within the individual, variability between individuals, and secular variability. 1.3.1.1 Measurement Variability Varying degrees of care can be exercised in selecting population samples, using measurement instruments, storing the measured data, and applying statistical treatments. Depending on the care taken during measurements, the resulting information may be quite variable. An exceptionally large standard deviation (see statistics later in this chapter) of a data set can be a warning signal. I. The Ergonomic Knowledge Base
Poor data
12
1. Size and Mobility of the Human Body
TABLE 1.1 Estimates of Average Stature (in cm) in 20 Regions of the Earth Region
Females
Males
Africa North West Southeastern
161 153 157
169 167 168
America, South/Central Native Indians European and African extraction
148 162
162 175
America, North
165
179
Asia North Southeast Near East
159 153 157
169 163 168
Australia European extraction
167
177
China South
152
166
Europe Northern Central Eastern Southeast France Iberian Peninsula
169 166 163 162 163 160
181 177 175 173 177 171
India Northern Southern
154 150
167 162
Japan
159
172
Source: Adapted from Jürgens, Aune, and Pieper (1990).
Changes over time
People differ
1.3.1.2 Intraindividual Variability An individual’s body size changes from youth to age, and it depends on nutrition, physical exercise, and health as well as genetics. Such changes in the same individual become apparent in longitudinal studies, in which an individual is observed over years and decades. Most (but not all) such changes with age follow the scheme shown in Fig. 1.6. During childhood and adolescence, body dimensions such as stature increase rapidly. From early 20s into the 50s, little change occurs and in general, stature remain almost steady. From the sixth decade on, many dimensions decline, while others—such as weight or bone circumference—often increase. 1.3.1.3 Interindividual Variability Individuals differ from each other in body proportions and in specific measurements such as arm length, weight, and height. Data describing a population sample are usually collected in a cross-sectional study, in which every subject is measured at about the same moment in time. This means that people of different ages, nutrition, fitness, etc., I. The Ergonomic Knowledge Base
13
1.3 AVAILABLE ANTHROPOMETRIC INFORMATION
TABLE 1.2 Measured Heights and Weights of Adults: Averages (with Standard Deviations when Available) Measured where, group (Date of publication)
Sample size
Stature (mm)
Weight (kg)
Algeria: Females (1990)
666
1576 (56)
61 (13)
Brazil: Males (1988)
3076
1699 (67)
nda
Cameroon: Urban Females (2006) Urban males, 35–44 years old (2006)
1156 558
1620 1721
64 75
about 600 about 600
1572 (53) 1705 (59)
52 (7) 67 (9)
France: Females (2006) Males (2006)
5510 3986
1625 (71) 1756 (77)
62 (12) 77 (13)
Germany: Female army applicants (2006) Male army applicants (2006)
301 1036
1674 1795
64 75
Great Britain: Females (2001) Males (2001)
3870 3233
1611 1746
68 81
India: Females (1997) Males (1997) East-Ctr. India male farm workers (2002) East India male farm workers (1997) South India male workers (1992) North-E. (Assam) Male farm workers (2016)
251 710 300 134 128 130
1523 (66) 1650 (70) 1638 (56) 1621 (58) 1607 (60) 1628 (46)
50 (10) 57 (11) 57 (7) 54 (7) 57 (5) 55 (7)
Iran: Female students (1997) Male students (1997)
74 105
1597 (58) 1725 (58)
56 (10) 66 (10)
Ireland: Males (1991)
164
1731 (58)
74 (9)
Italy: Females (1991) Females (2002) Males (1991) Males (2002)
753 386 913 410
1610 (64) 1611 (62) 1733 (71) 1736 (67)
58 (8) 58 (9) 75 (10) 73 (11)
Japan: Females (1990) Males (1990)
240 248
1584 (50) 1688 (55)
54 (6) 66 (8)
Netherlands: Females 18–65 years old (2002) Males 18–65 years old (2002)
691 564
1679 (75) 1813 (90)
73 (16) 84 (16)
Russia: Female herders (ethnic Asians) Female students (Russians) Female students (Uzbeks) Fem. factory workers (Russians) Fem. factory workers (Uzbeks) Male students (Russians) Male students (Uzbeks) Male factory workers (Russians) Male factory workers (ethnic mix) Male farm mechanics (Asians) Male coal miners (Russians) Male construction workers (Russians) (all 1999)
246 207 164 205 301 166 150 192 150 520 150 150
1588 (55) 1637 (57) 1578 (49) 1606 (53) 1580 (54) 1757 (56) 1700 (52) 1736 (61) 1700 (59) 1704 (58) 1801 (61) 1707 (69)
nda 61 (8) 56 (7) 61 (8) 58 (9) 71 (9) 65 (7) 72 (10) 68 (8) 64 (8) nda nda
China: Females (Taiwan) (2002) Males (Taiwan) (2002)
(Continued) I. The Ergonomic Knowledge Base
14
1. Size and Mobility of the Human Body
TABLE 1.2 Measured Heights and Weights of Adults: Averages (with Standard Deviations when Available) (cont.) Measured where, group (Date of publication)
Sample size
Stature (mm)
Weight (kg)
Saudi Arabia: Males (1986)
1440
1675 (61)
nda
Singapore: Male pilot trainees (1995)
832
1685 (53)
nda
Sri Lanka: Females (1985) Males (1985)
287 435
1523 (59) 1639 (63)
nda nda
Thailand: Females (1991) Females (1991) Males (1991) Males (1991)
250 711 250 1478
1512 (48) 1540 (50) 1607 (20) 1654 (59)
nda nda nda nda
Turkey: Male soldiers (1991)
5108
1702 (60)
63 (7)
about 3800 about 3800 2208 1986 1774 1475 4082 1264 1127 30 41
1625 1762 1629 (64) 1626 (64) 1756 (67) 1760 (nda) 1755 (67) 1640 (73) 1778 (79) 1559 (61) 1646 (60)
75 87 62 (8) 67 (11) 76 (11) 84 (nda) 85 (14) 69 (18) 86 (18) 49 59
U.S.A.: Females (2004) Males (2004) Female army soldiers (1989) Female army soldiers (2014) Male army soldiers (1989) Male army soldiers (2009) Male Army soldiers (2014) Canadian and US Females 18–26 years old (2002) Canadian and US Males 18–65 years old (2002) Vietnamese females living in the USA (1993) Vietnamese males living in the USA (1993)
Abbreviation: nda, no data available. Source: Adapted from Kroemer (2017) who listed all sources except Patel et al. (2016) for North-East (Assam) Indian farm workers.
FIGURE 1.6 Approximate changes in stature with age.
I. The Ergonomic Knowledge Base
15
1.3 AVAILABLE ANTHROPOMETRIC INFORMATION
are included in the sample set. Most textbooks, including this one, contain data gathered in cross-sectional studies. 1.3.1.4 Secular Variations Factual and anecdotal evidence shows that people in present day are larger, on average, than their ancestors; yet, reliable anthropometric information on this development is available only for about the last 100 years. During the latter half of the 20th century, stature increased in North America and in Europe by about 1 cm per decade, on average. The probable reason was that improved nutrition and hygiene have allowed individuals to achieve more of their genetically determined body size potential. If this explanation is correct, then the rate of increase should slowly taper off until a final body size is reached: indeed, for stature12 in the United States, this seems to happening. Also during the latter half of the 20th century, body weight increased by about 2 kg per decade, but since about 2010, that increase has risen dramatically and has generated concerns about obesity in many countries. Most reliable data depicting secular trends are from military surveys. However, soldiers constitute a selected sample of the general population: for example, they exclude people older than about 50 years as well as people who are unusual in their body dimensions, such as extremely short or tall. Further, only fairly healthy individuals are included.13
Trends taller and fatter
Military surveys
1.3.2 Population Samples The Human Biometry Data Bank at the University René Descartes in Paris contains extensive anthropometric information on Europeans. Publications of the World Health Organization (WHO) and ISO standards contain some international body size data. Unfortunately, sufficient information on most populations on Earth is missing. Available anthropometric data, such as listed in Table 1.2, are usually limited to a few dimensions, commonly stature and weight, often measured on small population samples. Body dimensions of soldiers have long been of interest for a variety of reasons, among them to provide uniforms, armor, and equipment. Armies have personnel willing to and capable of performing body measurements on large numbers of soldiers, available on command. Hence, anthropometric information about soldiers has a long history and is rather complete. For example, the US military anthropometric data bank contains the data of about 100 surveys from many nations, though most on US military personnel. Among the US military services, the Army is the largest and anthropometrically least biased sample of the total US adult population. Therefore, the body dimensions of the Army often serve as useful estimates for the general North American adult population. However, the user of this information must realize that the military consists of relatively young and healthy individuals. Thus, civilian data are likely to differ somewhat; only the soldiers’ heads, hands, and feet are considered similar to civilians’ dimensions.14 It is of some interest to note that even the military data show strong increases in body weight from 1988 to 2012,15 accompanied by enlarged trunk depth measures and associated larger circumferences especially of the trunk and thighs—apparently, the worldwide obesity trend pertains both to soldiers and civilians. Table 1.3 provides anthropometric data excerpted from surveys done on Chinese in Taiwan measured between 1996 and 2000; on Russians in Moscow measured between I. The Ergonomic Knowledge Base
Civilians’ body sizes
Body sizes of soldiers
US Army body sizes
Body sizes in China, Russia and United States
16
1. Size and Mobility of the Human Body
TABLE 1.3 Common Body Measurements and Their Applications Females mean (SD)
Males mean (SD)
Chinese (Taiwan) Russians (Moscow) US Army soldiers
1572 (53) 1637 (57) 1629 (64)
1705 (59) 1757 (56) 1756 (69)
2. Eye height, standing Origin of the visual field of a standing person. Reference for the location of visual obstructions and of visual targets such as displays; consider slump and motion
Chinese (Taiwan) Russians (Moscow) US Army soldiers
nda 1526 (57) 1520 (62)
nda 1637 (55) 1642 (67)
3. Shoulder height (acromion), standing Starting point for arm length measurements; near the center of rotation of the upper arm. Reference point for hand reaches; consider slump and motion
Chinese (Taiwan) Russians (Moscow) US Army soldiers
1285 (50) 1334 (54) 1335 (58)
1396 (53) 1440 (54) 1441 (63)
4. Elbow height, standing Reference for height and distance of the work area of the hand and the location of controls and fixtures; consider slump and motion
Chinese (Taiwan) Russians (Moscow) US Army soldiers
978 (38) 1010 (42) 1004 (45)
1059 (40) 1083 (48) 1083 (50)
5. Hip height (trochanter), standing Traditional anthropometric measure, indicator of leg length and the height of the hip joint. Used for comparing population samples
Chinese (Taiwan) Russians (Moscow) US Army soldiers
802 (41) nda 845 (45)
860 (48) nda 901 (49)
6. Knuckle height, standing Reference for low locations of controls, handles, and handrails; consider slump and motion of the standing person
Chinese (Taiwan) Russians (Moscow) US Army soldiers
708 (33) 731 (34) nda
757 (32) 773 (39) nda
7. Fingertip height (dactylion), standing Reference for the lowest locations of controls, handles, and handrails; consider slump and motion of the standing person
Chinese (Taiwan) Russians (Moscow) US Army soldiers
618 (32) 635 (32) 613 (35)
659 (30) 668 (37) 654 (37)
8. Sitting height Reference for the minimal height of overhead obstructions; add to height to allow for clearance, head covering, trunk motion of the seated person
Chinese (Taiwan) Russians (Moscow) US Army soldiers
846 (32) 859 (32) 857 (33)
910 (30) 912 (32) 918 (36)
9. Sitting eye height Origin of the visual field of a seated person. Reference point for the location of visual targets such as displays; consider slump and motion
Chinese (Taiwan) Russians (Moscow) US Army soldiers
732 (31) 742 (29) 748 (30)
791 (29) 790 (33) 805 (33)
10. Sitting shoulder height (acromion) Starting point for arm length measurements; near the center of rotation of the upper arm. Reference for hand reaches; consider slump and motion
Chinese (Taiwan) Russians (Moscow) US Army soldiers
561 (27) nda 563 (29)
602 (26) nda 603 (31)
11. Sitting elbow height Reference for the height of an armrest, of the work area of the hand and of keyboard and controls; consider slump and motion of the seated person
Chinese (Taiwan) Russians (Moscow) US Army soldiers
252 (25) 236 (24) 232 (26)
264 (24) 243 (25) 245 (29)
12. Sitting thigh height (clearance) Reference for the minimal clearance needed between seat pan and the underside of a table or desk; add clearance for clothing and motions
Chinese (Taiwan) Russians (Moscow) US Army soldiers
nda 148 (14) 168 (14)
nda 151 (18) 181 (16)
13. Sitting knee height Traditional anthropometric measure for lower leg length. Reference for the minimal clearance below the underside of a table or desk; add shoe height
Chinese (Taiwan) Russians (Moscow) US Army soldiers
471 (24 527 (24) 511 (27)
521 (29) 562 (25) 554 (28)
14. Sitting popliteal height Reference for the height of a seat; add shoe height
Chinese (Taiwan) Russians (Moscow) US Army soldiers
379 (18) 423 (23) 388 (24)
411 (19) 468 (24) 430 (25)
Dimensions, applications
Population
1. Stature A main measure for comparing population samples. Reference for the minimal height of overhead obstructions; add to height to allow for clearance, head covering, shoes, stride
I. The Ergonomic Knowledge Base
17
1.3 AVAILABLE ANTHROPOMETRIC INFORMATION
TABLE 1.3 Common Body Measurements and Their Applications (cont.) Females mean (SD)
Males mean (SD)
Chinese (Taiwan) Russians (Moscow) US Army soldiers
309 (18) nda 334 (17)
338 (19) nda 364 (18)
16. Elbow-fingertip length Traditional anthropometric measure. Reference for fingertip reach when moving the forearm in the elbow
Chinese (Taiwan) Russians (Moscow) US Army soldiers
384 (27) nda 440 (23)
427 (27) nda 480 (23)
17. Overhead grip reach, sitting Reference for the height of overhead controls operated by a seated person. Consider ease of motion, reach, and finger/hand/arm strength
Chinese (Taiwan) Russians (Moscow) US Army soldiers
1105 (44) 1169 (46) 1196 (62)
1208 (49) 1276 (47) 1303 (68)
18. Overhead grip reach, standing Reference for the height of overhead controls operated by a standing person. Add shoe height. Consider ease of motion, reach, and strength
Chinese (Taiwan) Russians (Moscow) US Army soldiers
1831 (67) nda 1968 (98)
2002 (79) nda 2141 (104)
19. Forward grip reach Reference for forward reach distance. Consider ease of motion, reach, and finger/ hand/arm strength
Chinese (Taiwan) Russians (Moscow) US Army soldiers
651 (33) 702 (37) 693 (43)
710 (36) 759 (38) 757 (44)
20. Arm length, vertical A traditional measure for comparing population samples. Reference for the location of controls very low on the operator’s side. Consider ease of motion, reach, strength
Chinese (Taiwan) Russians (Moscow) US Army soldiers
669 (31) nda 722 (37)
738 (33) nda 787 (39)
21. Downward grip reach Reference for the location of controls low on the side of the operator. Consider ease of motion, reach, and finger/hand/arm strength
Chinese (Taiwan) Russians (Moscow) US Army soldiers
nda nda 607 (30)
nda nda 663 (32)
22. Chest depth A traditional measure for comparing population samples. Reference for the clearance between seat backrest and the location of obstructions in front of the trunk
Chinese (Taiwan) Russians (Moscow) US Army soldiers
213 (19) 242 (21) 247 (27)
217 (19) 245 (20) 254 (26)
23. Abdominal depth, sitting A traditional measure for comparing population samples. Reference for the clearance between seat backrest and the location of obstructions in front of the trunk
Chinese (Taiwan) Russians (Moscow) US Army soldiers
nda nda 223 (32)
nda nda 255 (37)
24. Buttock-knee depth, sitting Reference for the clearance between seat backrest and the location of obstructions in front of the knees
Chinese (Taiwan) Russians (Moscow) US Army soldiers
530 (26) 584 (29) 591 (33)
558 (31) 610 (30) 618 (31)
25. Buttock-popliteal depth, sitting Reference for the depth of a seat.
Chinese (Taiwan) Russians (Moscow) US Army soldiers
nda 496 (29) 485 (29)
nda 517 (26) 503 (27)
26. Shoulder breadth (biacromial) Chinese (Taiwan) A traditional measure for comparing population samples. Indicator of the distance Russians (Moscow) between the centers of rotation of the two upper arms US Army soldiers
324 (25) 360 (16) 365 (18)
369 (28) 397 (25) 416 (19)
27. Shoulder breadth (bideltoid) Reference for the lateral clearance required near shoulder level. Add space for ease of motion and tool use
Chinese (Taiwan) Russians (Moscow) US Army soldiers
406 (24) 412 (21) 450 (29)
460 (23) 458 (23) 510 (33)
28. Hip breadth, sitting Reference for seat width. Add space for clothing and ease of motion
Chinese (Taiwan) Russians (Moscow) US Army soldiers
353 (23) 372 (23) 399 (33)
360 (27) 362 (23) 379 (30)
Dimensions, applications
Population
15. Shoulder-elbow length Traditional anthropometric measure for comparing population samples
(Continued)
I. The Ergonomic Knowledge Base
18
1. Size and Mobility of the Human Body
TABLE 1.3 Common Body Measurements and Their Applications (cont.) Females mean (SD)
Males mean (SD)
Chinese (Taiwan) Russians (Moscow) US Army soldiers
1571 (62) 1640 (75) 1660 (83)
1738 (69) 1782 (68) 1814 (85)
30. Elbow span (arms akimbo) Reference for the lateral space needed at upper body level for ease of motion and tool use
Chinese (Taiwan) Russians (Moscow) US Army soldiers
801 (39) 870 (38) nda
894 (45) 935 (37) nda
31. Head length (depth) A traditional measure for comparing population samples. Reference for headgear size
Chinese (Taiwan) Russians (Moscow) US Army soldiers
187 (6) nda 190 (7)
197 (7) nda 196 (7)
32. Head breadth A traditional measure for comparing population samples. Reference for headgear size
Chinese (Taiwan) Russians (Moscow) US Army soldiers
161 (9) nda 148 (5)
167 (8) nda 154 (6)
33. Hand length A traditional measure for comparing population samples. Reference for hand tool and gear size. Consider manipulations, gloves, tool use
Chinese (Taiwan) Russians (Moscow) US Army soldiers
167 (8) 168 (8) 181 (10)
183 (10) 188 (9) 193 (10)
34. Hand breadth A traditional measure for comparing population samples. Reference for hand tool and gear size, and for an opening through which a hand must fit. Consider gloves, tool use
Chinese (Taiwan) Russians (Moscow) US Army soldiers
75 (4) 76 (3) 78 (4)
86 (5) 87 (5) 88 (4)
35. Foot length A traditional measure for comparing population samples. Reference for shoe and pedal size
Chinese (Taiwan) Russians (Moscow) US Army soldiers
nda 239 (11) 246 (12)
nda 266 (12) 271 (13)
36. Foot breadth A traditional measure for comparing population samples. Reference for shoe size, spacing of pedals
Chinese (Taiwan) Russians (Moscow) US Army soldiers
nda 88 (4) 93 (5)
nda 97 (6) 102 (5)
37. Weight (in kg) Chinese (Taiwan) A traditional measure for comparing population samples. Reference for body size, cloth- Russians (Moscow) ing, strength, health, etc. Add weight for clothing and equipment worn on the body US Army soldiers
52 (7) 60 (7) 68 (11)
67 (9) 71 (9) 86 (14)
Dimensions, applications
Population
29. Span A traditional measure for comparing population samples. Reference for sideway reach
Dimensions correspond to numbering in Figs. 1.7–1.9. All measures represent the mean (with standard deviation in parentheses) in mm, except weight in kg. Abbreviation: nda, no data available. Gordon et al. (2014) provide exact definitions of the measurements. Sources: Chinese (Taiwan), 25–34 years of age, measured between 1996 and 2000: Wang, Wang and Lin (2002). Russians, students in Moscow, 18–22 years of age, measured between 1984 and 1986: Strokina and Pakhomova (1999). US Army soldiers, 4082 males and 1986 females, 17–58 years of age, measured 2010, 2011, and 2012: Gordon et al. (2014).
1984 and 1986; and on US Army soldiers measured between 2010 and 2012. Figs. 1.7–1.9 depict the tabulated body dimensions.
1.4 ANTHROPOMETRIC STATISTICS The mean m (often called average), standard deviation S, and sample size n completely describe a normally distributed (Gaussian, bell-shaped) data set.
I. The Ergonomic Knowledge Base
FIGURE 1.7 Measurements of heights; numbered as in Table 1.3.
FIGURE 1.8 Measurements of reaches, heights, depths, breadths, and spans; numbered as in Table 1.3.
20
1. Size and Mobility of the Human Body
FIGURE 1.9 Measurements taken on head, hand, and foot; numbered as in Table 1.3.
Normal distributions
Easy statistics
FIGURE 1.10 A representative normal distribution of stature measurements: 95% of all data points lie between 162 and 188 cm.
Conveniently, anthropometric data usually appear in a reasonably normal distribution, where all data points fall along a bell-shaped contour, symmetrical to the left and right of the average. (Unfortunately, there are some important exceptions: weight and muscle strength data are usually not normally distributed.) Fig. 1.10 sketches such a normal (or Gaussian) distribution where the measured heights appear in bell shape on both sides of the mean value, here about 175 cm. Most anthropometric data sets can be easily described by simple (parametric) statistics, listed in Table 1.4. The data cluster at the center of the set, symmetrically left and I. The Ergonomic Knowledge Base
21
1.4 ANTHROPOMETRIC STATISTICS
TABLE 1.4 Statistical Formulas of Particular Use in Anthropometry Measures of central tendency Mean, average m
m = ∑x/n (where n is the number of data points)
Median
Middle value of values in numerical order
Mode
Most often occurring value
E#1
Measures of variability Range
xmax − xmin
E#2
Standard deviation S
S = (variance)1/2 = [∑(x − m)2/n] 1/2
E#3
Coefficient of variation CV
CV = S/m (in percent: CV = 100S/m %)
E#4
Standard error of the mean SE
SE = S/n1/2
E#5
Symmetry (skewness) β1
β1 = ∑(x − m) /nS
Peakedness (kurtosis) β2
3
3
E#6
4
4
E#7
B2 = ∑(x − m) /nS
Percentile value location in a data distribution Percentile p
p = m + kS (k from Table 1.5)
E#8
Sample size n (for m)
n ≥ (1.96 CV)2 (CV in %)
E#9a
Sample size n (for 5th/95th p)
n ≥ (1.96 CV)2 (1.534)2 (CV in %)
E#9b
Sampling
Measures of relations between variables x and y Correlation coefficient r between two samples x and y
r = Sxy/(SxSy) r = ∑[(x – mx)(y - my)]/[∑(x − mx)2 ∑(y − my)2]1/2
E#10
Coefficient of determination R
R = r2
E#11
Regression
y = a + b x b = r Sy/Sx
E#12 E#13
E = dx/Dx = dy/Dy
E#14
Covariance COV
COVx,y = rxy Sx Sy
E#15
Sum of two mean values
mz = mx + my
Standard deviation of the sum
Sz = S + S + 2 rSx Sy
Difference of two mean values
mz = mx – my
Standard deviation of the difference
Sz = S + S − 2 rSx Sy
Factor k in a combined set
kz = nxkx + nyky
E#20
Percentile p in a combined set
pz = nxpx + nypy
E#21
Ratio scaling Scaling factor E Combining data sets
( (
2 x
2 x
2 y
2 y
E#16
)
1/2
E#17 E#18
)
1/2
E#19
right of the 50th percentile, which coincides with the mean m (the arithmetic average). The standard deviation S (or SD) describes the variability (spread) of all collected data; the sample size is n. Skewness β1 is dimensionless; if its value is zero, the set of data is symmetrically distributed. Kurtosis β2 is also dimensionless; if its value is 3, the data are normally distributed. (That value is set to zero in some other formulas.) I. The Ergonomic Knowledge Base
22 No “average person”
Check the CV
Use percentiles
1. Size and Mobility of the Human Body
It’s easy to calculate the average in a set of data, but that mean value is just one of the statistical descriptors of the distribution.16 One cannot “design for the average” because that design would be too big for half the people and too small for the other half—consider the mean headaches caused by a doorway at mean stature. Furthermore, practically nobody is average in many or all features. People differ, and ergonomic design must accommodate variability. One simple way to check on data diversity is to divide the standard deviation of the data in question by their mean to get the coefficient of variation (CV)—see E#4 in Table 1.4. For most body dimensions the CV is between 3% and 10%; larger values are suspect and should prompt a thorough examination of the data. (However, in most strength data the CV is between 10% and 85%.) The location of every datum in a distribution of data can be described by its percentile value. For example: in Fig. 1.10, the 2.5th percentile is at 162 cm, so 2.5% of all data are smaller while 97.5% are larger; the range from 162 to 188 cm includes 95% of all data. Accommodating specific body dimensions, or a range of body sizes, is best done by selecting percentile points in sets of anthropometric data. Only judicious use of percentiles avoids the fallacy of “designing for the average.”
P E R C E N T I L E S A R E U S E F U L I N S E V E R A L WAY S First, they help to establish the portion of a user population that will be included in (or excluded from) a specific design solution. For example: a certain product may need to fit everybody who is larger than 5th percentile or smaller than 95th percentile in a specified dimension, such as grip size or arm reach. So, the central 90% of all users will be accommodated, but the 5% having values smaller than 5th percentile and the 5% having values larger than 95th percentile will not be fitted. This “5 to 95” strategy is widely used. Second, percentiles help to select subjects for fit trials. For example: if the product needs to be tested, individuals having 5th or 95th percentile values (or any other chosen values) in the critical dimensions are selected to participate in use trials. Third, any body dimension, design value, or subject score can be exactly located in the general distribution. For example: a certain foot length can be identified as a given percentile value of that dimension; a given seat height can be describe as fitting a distinct percentile value of lower leg length (popliteal height); a test score can be explained as being a discrete percentile value on the distribution of scores. Finally, the use of percentiles helps in the selection of people who can use a given product. For example: if a cockpit of an airplane is designed to fit 10th–90th percentiles, one can select cockpit crews whose body measures are above the 10th and below the 90th percentile in the critical design dimensions.
Often, one wants to know a given percentile value, say the 5th percentile. There are two ways to determine this. One, check a graph of the data distribution, such as shown
I. The Ergonomic Knowledge Base
23
1.5 USING ANTHROPOMETRIC DATA
FIGURE 1.11 Skewed distribution sectioned into percentile portions.
in Fig. 1.11 and measure, count, estimate, or eyeball the critical value. This works whether the distribution is normal, skewed, binomial, or in any other form. If the data are normally distributed, there is a second, easy (and exact) approach: calculate. Calculation of a percentile values p uses Eq. (E#8) (see Table 1.4):
p = m + kS
(1.1)
with m the mean and S the standard deviation; take factor k from Table 1.5.
1.5 USING ANTHROPOMETRIC DATA Ergonomists need to use relevant and reliable body dimensions to design equipment and tools that utilize and enhance human strengths. Some specific examples follow.17
EXAMPLE: ARM LENGTH The task: Calculate 95p shoulder-to-fingertip length. The solution: You know the mean lower arm (LA) link length (with the hand) to be 443 mm with a standard deviation of 23 mm. The mean upper arm (UA) link length is 336 mm and its standard deviation is 17 mm. You are tempted to calculate both 95th percentile values and add them. But you may not use the sum of the two 95p lengths because this would disregard their correlation. Therefore, you must calculate the sum of the mean values first: mA = mLA + mUA = 443 + 336 = 779 mm (Equation E#16 in Table 1.4) Next, you calculate standard deviation of the new mean, using an assumed correlation coefficient of 0.4: 1/2 SDA = 232 + 17 2 + 2 × 0.4 × 23 × 17 = 33.6 mm (E#17) Now you can calculate the 95p total arm length: A 95 = 779 mm + 1.64 × 33.6 mm = 834 mm (E#8, with k = 1.64 from Table 1.5) The result: The 95p shoulder-to-fingertip length is calculated to be 834 mm.
I. The Ergonomic Knowledge Base
24
1. Size and Mobility of the Human Body
TABLE 1.5 Percentile Values and Associated k Factors Below mean Percentile
Above mean
Factor k
Percentile
Factor k
Percentile
0.001
−4.25
24
−0.71
50 (mean)
0.01
−3.72
25
−0.67
0.1
−3.09
26
−0.64
0.5
−2.58
27
1
−2.33
28
2
Factor k
Percentile
Factor k
0
77
0.74
51
0.03
78
0.77
52
0.05
79
0.81
−0.61
53
0.08
80
0.84
−0.58
54
0.10
81
0.88
−2.05
29
−0.55
55
0.13
82
0.92
2.5
−1.96
30
−0.52
56
0.15
83
0.95
3
−1.88
31
−0.50
57
0.18
84
0.99
4
−1.75
32
−0.47
58
0.20
85
1.04
5
−1.64
33
−0.44
59
0.23
86
1.08
6
−1.55
34
−0.41
60
0.25
87
1.13
7
−1.48
35
−0.39
61
0.28
88
1.18
8
−1.41
36
−0.36
62
0.31
89
1.23
9
−1.34
37
−0.33
63
0.33
90
1.28
10
−1.28
38
−0.31
64
0.36
91
1.34
11
−1.23
39
−0.28
65
0.39
92
1.41
12
−1.18
40
−0.25
66
0.41
93
1.48
13
−1.13
41
−0.23
67
0.44
94
1.55
14
−1.08
42
−0.20
68
0.47
95
1.64
15
−1.04
43
−0.18
69
0.50
96
1.75
16
−0.99
44
−0.15
70
0.52
97
1.88
17
−0.95
45
−0.13
71
0.55
97.5
1.96
18
−0.92
46
−0.10
72
0.58
98
2.05
19
−0.88
47
−0.08
73
0.61
99
2.33
20
−0.84
48
−0.05
74
0.64
99.5
2.58
21
−0.81
49
−0.03
75
0.67
99.9
3.09
22
−0.77
50 (mean)
0
76
0.71
99.99
3.72
23
−0.74
99.999
4.26
Any percentile value p (in a normal distribution of data) can be calculated from the mean m and the standard deviation S, p = m + k S. (This is E#8 in Table 1.4.) Note that k is negative for percentile values below the mean.
I. The Ergonomic Knowledge Base
25
1.6 BODY PROPORTIONS
EXAMPLE: TORSO MASS The task: Determine the mass of the female torso at 75th percentile. The solution: From a recent publication, you take it that the mass of torso and head combined has an average of 35.8 kg and a standard deviation of 5.2 kg. The estimated mass of the head, measured separately, has a mean of 5.8 kg and a standard deviation of 1.2 kg. You assume the correlation between head and torso masses to be 0.1. The mean torso mass is the difference between the average values of torso and head masses: mtorso = 35.8 kg − 5.8 kg = 30.0 kg (Equation E#18 in Table 1.4) The standard deviation of the mean torso mass results from: 1/2 SDtorso = 5.2 2 + 1.2 2 − 2 × 0.1 × 5.2 × 1.2 kg = 5.2 kg (E#19) The mass of a 75th percentile torso is: mass torso 75p = 30.0 kg + 0.67 × 5.2 kg = 33.5 kg (E#8, with k = 0.67 from Table 1.5) The result: The mass of the 75th percentile torso is calculated to be 33.5 kg.
1.6 BODY PROPORTIONS We often judge the human body by how its components “fit together.” European images of a beautiful person are affected by esthetic codes, canons, and rules founded on often ancient (Egyptian, Greek, Roman) beauty concepts of the human body. Leonardo da Vinci’s drawing of the body within a frame of graduated circles and squares has been adopted, in simplified form, as the emblem of the US Human Factors and Ergonomics Society. Categorizing body builds into different types is called somatotyping (from the Greek soma for body). In the 1940s, William Herbert Sheldon established a system of three body types, intended to describe (male) body proportions. Sheldon rated each person’s appearance in proportions of ectomorphic, endomorphic, and mesomorphic components: long/lean/slim, strong/sturdy/large, and stocky/solid/muscular body builds, respectively. This typology was originally based on intuitive assessment, not on actual body measurements; these were introduced into the system later by Sheldon’s disciples. In 1967, Heath and Carter standardized the procedure using Sheldon’s terms: their body typology has been widely employed. Unfortunately, these and other attempts at somatotyping have not provided reliable predictors of human performance in technological systems and, hence, are of little value for ergonomists.18
The body beautiful
Types of body build
1.6.1 Body Image Everyone has a mental picture of the physical appearance of the own body. This body image may affect lifestyle behaviors, in particular when we have a distorted view of our own appearance: a common associate of some weight and eating problems, especially in
I. The Ergonomic Knowledge Base
How we see ourselves
26
1. Size and Mobility of the Human Body
societies that scorn the obese. For example, a person suffering from anorexia nervosa may appear emaciated yet still complain of looking and feeling “fat.” Individuals who engage intensely in athletics requiring rigid weight or body shape limitations, such as ballet dancers, gymnasts, and runners, are prone to exhibit body image disturbances and may exercise compulsively and in excess.
ASKING ABOUT BODY SIZE INSTEAD OF MEASURING? Self-reporting is notoriously inaccurate: women and men tend to underreport their weight,19 short people tend to overestimate their stature while heavy people often underestimate their weight.
1.6.2 “Desirable” Body Weight Healthy body weight?
Adipose (fat-containing) tissue is a normal part of the human body: it stores fat for use as energy under high metabolic demands. However, excess of such fatty tissue is an obvious indicator of obesity. People who are severely overweight have a higher risk of health problems and of early death than their slimmer contemporaries do: the more overweight, the higher the risk. Causes for obesity may be both genetic and behavioral: they include too much caloric intake, too little physical activity, and metabolic and endocrine malfunctions.
1.6.3 Body Mass Index BMI
It is a matter of judgment (since there are no natural cutoff points) to establish quantitative definitions of normal body weight, of emaciation or underweight compared to overweight and obesity. Adolphe Quetelet (1796–1874) calculated a Body Mass Index (BMI) by dividing a person’s body weight (in kilograms) by the square of body height (stature, in meters). Quetelet’s BMI procedure has seen widespread use since the 1990s. The WHO classifies BMI values below 18.5 as “underweight,” above 25 as “overweight,” and BMI values above 30 as “obese.”
Using the metric system, to calculate BMI, divide weight (kg) by squared height (m). BMI = (weight [kg])/(height [m])2 Example: if height = 1.65 m and weight = 68 kg, then BMI = 68/(1.65)2 = 25 Using the old US system, to calculate BMI, divide weight (lbs) by squared height (in), then multiply with a conversion factor of 703. BMI = 703 × (weight [lb])/(height [in])2 Example: if height = 5′5″ = 65 in and weight = 150 lbs, then BMI = [150/(65)2] × 703 = 25
I. The Ergonomic Knowledge Base
27
1.7 DEALING WITH STATISTICS
Among the troublesome issues with the concept of BMI20 is that body composition varies among individuals of the same age, height, and weight. Age is an important moderator of the BMI-to-fatness relationship; at equal BMIs, older individuals are fatter than younger adults. BMI varies strongly with age in children. As women generally have smaller bones and less muscle tissue than men, one might expect that women’s BMIs would be less than those of men for any given percentile; however, this generalization actually applies only below the 75th percentile of the BMI distribution. In the upper quarter of the BMI distribution, women’s BMIs are generally higher than men’s. (See also Chapters 3 and 14 for additional considerations of BMI.)21
Troubling BMI issues
1.7 DEALING WITH STATISTICS Some groups of anthropometric data show strong covariation with each other: as one increases, another (or several others) increases as well—this happens, for example, among weight and trunk circumferences. Conversely, as one measure increases, others may decrease: as we advance into old age, often body heights decrease (see Fig. 1.6). Many body dimensions show little correlation with each other. Fig. 1.12 shows examples of how values of one variable can scatter with respect to another variable; the relationships may be positive or negative, strong or weak. The correlation coefficient, r, (formally, the Pearson product–moment correlation coefficient; E#10 in Table 1.4) numerically expresses the linear relationship between two variables. Human stature does not correlate well with weight (or head length, or waist circumference, and other measures), whereas other body dimensions are closely related, such as eye height with stature. Table 1.6A lists selected correlation coefficients r among body dimensions of US Army personnel, male and female.22 Unfortunately, similar tables for other nationalities and for civilians seem not to exist. The data in Table 1.6B show that even in the military, a group of individuals selected for their youthful fitness, there is little correlation between stature and weight: r is only around 0.5. This points to a basic problem with the BMI and with other body indices which rely on the presumption of a robust positive relation between body weight and stature—which does not exist. Unfortunately, some people still believe in an ill-conceived 1966 design scheme which presented all body heights, all body breadths, and all segment lengths as fixed percentages of stature. For instance, hip breadth was said to be 19.1% of height—misleading nonsense, of course, because hip breadth relates well only to a few trunk circumferences but varies widely among individuals and between males and females as groups. Furthermore, what useful gadget could be designed for a fixed “average” hip breadth? The linear relationship between a dependent variable y and an independent variable x is a bivariate regression as expressed by the Eq. (y = a + bx) where a is the intercept and b the slope (E#12 in Table 1.4). Note that often a linear relationship between x and y is assumed but not verified. The coefficient of determination, R, is simply the square of the correlation coefficient r between the two variables used in a bivariate regression equation (or among more variables in multiple regression equations). R measures the strength of association by
I. The Ergonomic Knowledge Base
Relations among body dimensions
Anthropometric correlations
Lacking covariation
Regression
Determination
28
1. Size and Mobility of the Human Body
FIGURE 1.12 Scatter diagrams and associated correlations.
“0.7 Convention”
stating the proportion of variation in the dependent variable y which is predictable from the independent variable x. It is common practice in engineering anthropometry (in fact, in ergonomics altogether) to require a correlation coefficient r of at least 0.7 as a basis for design decisions that involve related body dimensions. The reason for this” 0.7 convention” is that one should be able to explain at least 50% of the variance of the predicted value from of the predictor
I. The Ergonomic Knowledge Base
TABLE 1.6 Correlations between Anthropometric Data on US Soldiers 1 Age
3 Stat
4 OFR
5 WH
6 CH
7 SH
8 PH
9 SC
10 CC
0.219
0.041
0.017
0.044
−0.055
0.066
−0.074
0.155
0.193
0.529
0.493
0.491
0.370
0.422
0.242
0.845
0.806
0.928
0.848
0.840
0.755
0.808
0.377
0.222
0.704
0.905
0.554
0.868
0.384
0.199
0.625
0.754
0.587
0.300
0.255
0.330
0.915
0.267
0.093
0.343
0.285
0.202
0.188
0.023
1
Age [302]
2
Weight [125]
0.195
3
Stature [100]
−0.021
0.546
4
Overhead fingertip reach [84]
−0.013
0.525
0.937
5
Wrist height, standing [128]
0.028
0.527
0.856
0.749
6
Crotch height [39]
0.090
0.351
0.852
0.890
0.673
7
Sitting height [94]
0.026
0.447
0.741
0.578
0.692
0.347
8
Popliteal height, sitting [87]
−0.094
0.341
0.852
0.883
0.673
0.924
0.383
9
Shoulder circumference [91]
0.122
0.861
0.399
0.413
0.334
0.250
0.326
0.256
10
Chest circumference [34]
0.279
0.873
0.312
0.308
0.357
0.135
0.287
0.137
0.859
11
Waist circumference [115]
0.364
0.849
0.276
0.251
0.343
0.060
0.298
0.074
0.703
0.839
12
Buttock circumference [24]
0.190
0.935
0.401
0.380
0.412
0.204
0.373
0.191
0.781
0.815
13
Span [99]
−0.016
0.497
0.815
0.908
0.535
0.840
0.398
0.844
0.445
0.281
14
Biacromial breadth [11]
0.034
0.496
0.487
0.506
0.295
0.370
0.407
0.394
0.633
0.419
15
Hip breadth, standing [66]
0.209
0.831
0.453
0.416
0.457
0.239
0.464
0.224
0.672
0.727
16
Head circumference [62]
0.125
0.508
0.342
0.312
0.302
0.224
0.303
0.240
0.433
0.421
17
Head length [63]
−0.002
0.371
0.346
0.315
0.295
0.260
0.302
0.268
0.295
0.271
18
Head breadth [61]
0.198
0.320
0.114
0.098
0.112
0.034
0.128
0.035
0.303
0.311
19
Hand length [60]
0.032
0.453
0.650
0.724
0.464
0.676
0.300
0.679
0.372
0.242
20
Foot length [52]
−0.012
0.512
0.700
0.734
0.537
0.687
0.383
0.697
0.409
0.299
0.808
1.7 DEALING WITH STATISTICS
I. The Ergonomic Knowledge Base
2W
(Continued)
29
30
TABLE 1.6 Correlations between Anthropometric Data on US Soldiers (cont.) 11 WC
12 BC
13 Sp
14 BB
15 HiB
16 HC
17 HeL
18 HeB
19 HaL
20 FL
Age [302]
0.299
0.258
0.011
0.025
0.283
0.073
0.027
0.044
0.044
0.026
2
Weight [125]
0.767
0.897
0.438
0.440
0.778
0.428
0.329
0.420
0.430
0.493
3
Stature [100]
0.167
0.361
0.787
0.505
0.372
0.348
0.354
0.124
0.637
0.673
4
Overhead fingertip reach[84]
0.132
0.313
0.907
0.535
0.294
0.337
0.345
0.095
0.737
0.732
5
Wrist height, standing [128]
0.217
0.363
0.453
0.303
0.397
0.250
0.261
0.403
0.403
0.468
6
Crotch height [39]
0.061
0.185
0.870
0.418
0.146
0.287
0.302
0.043
0.706
0.703
7
Sitting height [94]
0.142
0.351
0.336
0.384
0.438
0.246
0.255
0.159
0.256
0.330
8
Popliteal height, sitting [87]
−0.031
0.063
0.840
0.420
0.051
0.241
0.271
0.020
0.685
0.671
9
Shoulder circumference [91]
0.697
0.726
0.395
0.574
0.601
0.353
0.264
0.261
0.355
0.379
10
Chest circumference [34]
0.781
0.707
0.167
0.304
0.603
0.393
0.191
0.246
0.186
0.288
11
Waist circumference [115]
0.738
0.109
0.214
0.673
0.223
0.117
0.229
0.127
0.170
12
Buttock circumference [24]
0.859
0.258
0.327
0.915
0.313
0.226
0.220
0.258
0.323
13
Span [99]
0.201
0.352
0.565
0.203
0.345
0.338
0.083
0.827
0.775
14
Biacromial breadth [11]
0.311
0.411
0.575
0.294
0.287
0.259
0.152
0.441
0.456
15
Hip breadth, standing [66]
0.799
0.902
0.355
0.404
0.232
0.160
0.196
0.180
0.250
16
Head circumference [62]
0.376
0.427
0.320
0.301
0.364
0.824
0.497
0.342
0.360
17
Head length [63]
0.222
0.301
0.304
0.235
0.259
0.820
0.131
0.337
0.339
18
Head breadth [61]
0.277
0.268
0.131
0.180
0.235
0.541
0.120
0.082
0.113
19
Hand length [60]
0.166
0.320
0.810
0.433
0.298
0.330
0.306
0.137
20
Foot length [52]
0.220
0.390
0.766
0.445
0.377
0.333
0.304
0.161
0.825 0.806
Values for women are listed above the diagonal, for men below. Values larger than 0.7 are bolded. Pairs of data that correlate above 0.700 are similar in the male and female groups; also, correlations below 0.300 are similar for both genders.
1. Size and Mobility of the Human Body
I. The Ergonomic Knowledge Base
1
31
1.8 HOW TO OBTAIN MISSING DATA
variable: This requires R = r2 to be at least 0.5, so r is at least 0.7075. (Note that r depends on the sample size n).
1.8 HOW TO OBTAIN MISSING DATA In terms of anthropometry, Europe and North America long had the best-measured populations on earth,23 now joined by Russians and Taiwanese—see Table 1.2. Yet even there, most civilian populations were not assessed comprehensively, and current data on subgroups are sparse. Lucky is the ergonomist who is particularly interested in body sizes of Italians visiting swimming beaches,24 Portuguese and American children,25 Irish workers,26 Indian vehicle drivers,27 American farmers,28 pregnant American women,29 American wheelchair users,30 or of hand sizes.31 But in many cases, the exact body dimensions needed for a design are neither available in the literature nor on the internet. Occasionally, one can simply use the data of a population of known dimensions, if one has good reason to assume that this population is similar to the one on which data are missing. Yet one has to consider the underlying assumption: for example, are Taiwanese similar in size to all Chinese? Do Japanese males have body proportions that are similar to those of their American contemporaries?32 Are the body sizes of Russian students in Moscow similar to those of students in Tashkent? Can American women’s body sizes be treated as small men’s sizes33 for equipment design purposes? For a quick check, it may suffice to take measures on a few coworkers to get a rough estimate. How much effort to put into the data depends on the importance of the application and the risk and consequences of being wrong. The most thorough approach is to actually measure a sufficiently large and carefully selected, statistically appropriate sample of the population to be fitted. The literature provides help in understanding important aspects of anthropometric surveys34 such as sampling composite populations, and, of course, of measuring procedures. Getting to understand such issues quickly leads to the conclusion that planning and conducting an anthropometric survey is a complex and time-consuming task which should be done by qualified anthropometrists and trained specialists. Say you are designing student desks and want to know how tall 10-year old girls who attend schools in Spain are. It is impossible to measure them all, so the true value of that population statistic (in this example, the girls’ mean height) cannot be observed and remains unknown. However, it is possible to estimate the true value, with some degree of certainty, from measures taken on a random sample from the whole population of interest. In order to trust the sample statistics, the sampling procedure needs to be carefully planned and executed: this is often a daunting, time-consuming, and expensive task. To determine the required sample size, one starts with a “worst case” scenario: select the dimension that is expected to show the largest CV (waist circumference, for example is highly variable); then decide whether the data should describe low/high percentile values (often the 5th/95th percentiles) or the mean value of the distribution. The sample size needed to obtain information at the tails of distributions is larger than needed to estimate the center of the distribution. When the precision is 1% and the confidence level is 95% (as often desired), the minimal sample sizes35 for every subgroup
I. The Ergonomic Knowledge Base
Estimate or calculate?
Measure?
Sampling
Sample size
32
Predict?
1. Size and Mobility of the Human Body
are n ≥ (1.96CV)2 (1.534)2 for estimation of 5th/95th percentiles, or n ≥ (1.96CV)2 to estimate mean values (E#9 in Table 1.4) with the CV in percent. A rather challenging task is the prediction of future body dimensions, needed when equipment must be designed for use in decades to come. In the 1960s, for example, NASA was concerned about the body sizes of astronauts by the end of the century. Fig. 1.13 shows predicted values for male NASA flight crews in terms of stature; predictions were also made for female astronauts. For both groups, stature was assumed to increase linearly over the years, but that “secular” growth has slowed since the turn of the century. Note that at that time, all astronauts were fairly small and short so that they could fit into the tiny cockpits of the early space vehicles.
Clothing sizing in the USA is typical for use, misuse, and nonuse of correlations. Sizing of clothes for men is a fairly well organized and standardized procedure. Most men’s jacket sizes run from 38 to 56, meaning that they should fit men with chest circumferences between 38 and 56 in, with increments of 1 or 2 in. So, chest circumference is the primary “predictor variable” for other size variables, such as coat length, shoulder width, and sleeve length. Similarly, slacks are ordered by waist circumference and shirts by neck circumference. In men’s shirts, a given neck circumference usually is associated with a given chest circumference, while sleeve length may vary by 1 or 2 in increments. This is an attempt to cover various body dimensions with a few shirt sizes, but it has obvious shortcomings: if a person needs a large neck size, the corresponding shirt also usually comes with ballooning chest and waist circumferences, which the wearer may not need. In contrast, women’s clothes are not well standardized. There was one ill-defined prototype size 12 (based mostly on 1941 measurements), from which larger and smaller sizes were derived in nonstandard manners, as deemed suitable by each manufacturer. Hence, a woman well fitted by clothes of size 10 made by one producer may need a size 12, or 8, in clothing tailored by another company. There remains a plethora of nonstandard sizing schemes (mis)labeled junior, misses, mature, petite, tall, and/ or plus sizing. In recent years, so-called vanity sizing or size inflation has complicated women’s clothing: ready-to-wear clothing of the same nominal size may now fit a larger body. This makes the wearer “feel better” about (or more likely to purchase) the size being considered.
The wide diversion of body dimensions has long been demonstrated but is often overlooked: Table 1.7 lists that in a sample of 4063 men, 1055 had a “middling” stature between 38th and 62nd percentile, this 26% spread being indeed a very generous approximation of “middle.” Selecting those who were also “middling” in other body measures showed that after the fourth iteration less than 1% of the remaining men were “in the middle.” Similarly, Table 1.8 shows that trying to accommodate the center range
I. The Ergonomic Knowledge Base
33
1.8 HOW TO OBTAIN MISSING DATA
FIGURE 1.13 Predictions of average stature of NASA male personnel. Source: Adapted from Roebuck et al. (1988).
TABLE 1.7 How many Individuals are “Approximately Middle” in Body Dimensions? Step
Sample size
Remaining in the “approximately middle” sample
1
Of the original 4063 men
1055 were between 38th and 62nd percentile in Stature
26%
2
Of the remaining 1055 men
302 were between 38th and 62nd percentile in Chest Circumference
7%
3
Of the remaining 302 men
143 were between 38th and 62nd percentile in Arm (sleeve) Length
4%
4
Of the remaining 143 men
73 were between 38th and 62nd percentile in Crotch Height
2%
5
Of the remaining 73 men
28 were between 38th and 62nd percentile in Torso Circumference